Re: please... read this ... it is strange...!!!

[kings]

#include<iostream.h>
#include<conio.h>
void main()
{
int a=0xf0f0;
cout<<sizeof(a)<<endl; // if your compiler is vc++, the result is 4byte.
cout<<hex<<a; //so the result a is f0f0.
}

the result
2
fffff0f0
i had compiled this program with tc3.0(dos version)or tcwin3.1
the result is same...
although sizeof(a) is 2byte,the result is 4byte...
i know your meaning about sign expending... i want to know definite reason
i guess it is related with hardware(bus). i'll expecting your answer~!!^^*



"rt11guru" <rt11guru@yahoo.com> wrote:
>
>"kings" <donggl278@hanmail.net> wrote:
>>
>>i'm using turbo c++3.0(dos version)
>>
>>#include<iostream.h>
>>#include<conio.h>
>>void main()
>>{
>> int a=0xf0f0;
>> cout<<hex<<a;
>>}
>>
>>
>>i'm expecting the result "f0f0" but the result is "fffff0f0"
>>i don't understanding...// help me please...
>
>It looks to me like you compiler declares an int to be 16 bits, while your
>cout is expecting output in hex format to be 32 bits. You getting a conversion
>to 32 bits with the sign bit extended.
>Like "Gerd" I'm using VC++ 6.0 and can't duplicate your result directly.
>However, the code
>
> short a = 0xf0f0;
> long b = a;
> cout<<hex<<b;
>
>yields your result.
>
>The reason it doesn't work directly on VC++, is that VC++ defines an int
>to be 32 bits, so
>int a = 0xf0f0;
>is the same as
>int a = 0x0000f0f0;
>

[Chris]

Hi,

It's really not that strange given the fact that you are most likely working
on a 32bit comp. Further, there should be no question as to what's happening
given the response that 'rt11guru' posted. As for a definite reason - the
sign bit is being extended through bits 16-31. You could also check this
by trying values like 0x00F0, 0x10F0, ... , 0x70F0 where the most significant
bit(15) is 0.

Chris


"kings" <donggl278@hanmail.net> wrote:
>
>#include<iostream.h>
>#include<conio.h>
>void main()
>{
> int a=0xf0f0;
> cout<<sizeof(a)<<endl; // if your compiler is vc++, the result is 4byte.
> cout<<hex<<a; //so the result a is f0f0.
>}
>
>the result
>2
>fffff0f0
>i had compiled this program with tc3.0(dos version)or tcwin3.1
>the result is same...
>although sizeof(a) is 2byte,the result is 4byte...
>i know your meaning about sign expending... i want to know definite reason
>i guess it is related with hardware(bus). i'll expecting your answer~!!^^*
>
>
>
>"rt11guru" <rt11guru@yahoo.com> wrote:
>>
>>"kings" <donggl278@hanmail.net> wrote:
>>>
>>>i'm using turbo c++3.0(dos version)
>>>
>>>#include<iostream.h>
>>>#include<conio.h>
>>>void main()
>>>{
>>> int a=0xf0f0;
>>> cout<<hex<<a;
>>>}
>>>
>>>
>>>i'm expecting the result "f0f0" but the result is "fffff0f0"
>>>i don't understanding...// help me please...
>>
>>It looks to me like you compiler declares an int to be 16 bits, while your
>>cout is expecting output in hex format to be 32 bits. You getting a conversion
>>to 32 bits with the sign bit extended.
>>Like "Gerd" I'm using VC++ 6.0 and can't duplicate your result directly.
>>However, the code
>>
>> short a = 0xf0f0;
>> long b = a;
>> cout<<hex<<b;
>>
>>yields your result.
>>
>>The reason it doesn't work directly on VC++, is that VC++ defines an int
>>to be 32 bits, so
>>int a = 0xf0f0;
>>is the same as
>>int a = 0x0000f0f0;
>>
>